Decided to make a quick video showing off some Xenogears combat with PGXP and xBRZ enabled. 60fps because it's a 60fps game, it looks stunning and I swear, one of the best looking games I've seen on the PlayStation! =)
(Audio is normal, not like last time)
Posts by unreal676
1 - 20 of 22 Posts
@TheDimensioner Oh wow, playing on a CRT too! Is the shaking really noticeable on the real hardware? I know it's a bit obvious on an LCD panel, also doesn't tend to look as good their either.
I decided to quickly test out FF IX and I don't seem to have any issue with xBRZ enabled for this game, I only tested really quickly, right up to when you meet Vivi, but that takes two cutscenes and neither of them made a static white screen appear for me.... I imagine something is wrong on your and not entirely the programs fault with this problem. I had xBRZ at 3x for my testing, but I don't think that'll matter.
I decided to play a few games to showcase PGXP as well, Blasto!, MediEvil and Jersey Devil (with Metal Gear Solid on the way maybe later tonight or tomorrow).
Blasto! may be one of the most difficult games to play on an emulator with high resolution/quality settings. I wager it'd bring most computers that aren't truly modern or run nice intel processors to their knees, as it did with mine. The slowdowns get worse as the game goes on because the fully 3D environments take up more screen space and get complex too....
MediEvil is a great game and has barely any slowdowns, looks great and sounds great. CPU Overclocking breaks this game.
Jersey Devil is another fun one, I call it platforming hell because it can be really difficult with some jumps that you should clearly be able to make........... Looks amazing though.
Blasto breaks in CPU+Memory mode, so don't use that.
MediEvil was fine I believe, also Vertex Caching worked without issues.
Jersey Devil I can't remember too much, but CPU overclocking is allowed, and I had Memory Mode with no Vertex Caching, because I think vertex caching tends to break 2D elements often.
Blasto -
Jersey Devil -
MediEvil -
Audio might be a little loud, but nothing "too" bad, sorry in advance though.
Edit: Oh, since we're talking about recording settings, I too use Fraps, to make things easier, disable Force RGB Loseless Recording, that setting is a killer. My Jersey Devil video of 15 minutes came out to 33GB of data..... 33!!!!!!! I had to re-encode it to a much more sane 1 GB or something around there, but that's just silly.
I record at 1360*1020, keeps the 4:3 aspect ratio (I play all games like that). I don't think it affects me too much when I record in either 30 or 60fps.
Edit x2: Won't showcase Metal Gear Solid, I can play with high settings, but enabling xBRZ makes the water scenes run abysmally slow =(. Game looks gorgeous.... maybe I'll just show it with the standard PlayStation styled textures going on.......
Tip for those playing MGS, CPU+Memory Logic mode seems to be the best way to make everything nice and stable, but it is a bit of a performance sink.
I typically play games that I don't use xBRZ in at 8x8 in ResHack, it seems to be a bit too much for Metal Gear Solid with CPU+Memory Logic mode.
Blasto! may be one of the most difficult games to play on an emulator with high resolution/quality settings. I wager it'd bring most computers that aren't truly modern or run nice intel processors to their knees, as it did with mine. The slowdowns get worse as the game goes on because the fully 3D environments take up more screen space and get complex too....
MediEvil is a great game and has barely any slowdowns, looks great and sounds great. CPU Overclocking breaks this game.
Jersey Devil is another fun one, I call it platforming hell because it can be really difficult with some jumps that you should clearly be able to make........... Looks amazing though.
Blasto breaks in CPU+Memory mode, so don't use that.
MediEvil was fine I believe, also Vertex Caching worked without issues.
Jersey Devil I can't remember too much, but CPU overclocking is allowed, and I had Memory Mode with no Vertex Caching, because I think vertex caching tends to break 2D elements often.
Blasto -
Jersey Devil -
MediEvil -
Audio might be a little loud, but nothing "too" bad, sorry in advance though.
Edit: Oh, since we're talking about recording settings, I too use Fraps, to make things easier, disable Force RGB Loseless Recording, that setting is a killer. My Jersey Devil video of 15 minutes came out to 33GB of data..... 33!!!!!!! I had to re-encode it to a much more sane 1 GB or something around there, but that's just silly.
I record at 1360*1020, keeps the 4:3 aspect ratio (I play all games like that). I don't think it affects me too much when I record in either 30 or 60fps.
Edit x2: Won't showcase Metal Gear Solid, I can play with high settings, but enabling xBRZ makes the water scenes run abysmally slow =(. Game looks gorgeous.... maybe I'll just show it with the standard PlayStation styled textures going on.......
Tip for those playing MGS, CPU+Memory Logic mode seems to be the best way to make everything nice and stable, but it is a bit of a performance sink.
I typically play games that I don't use xBRZ in at 8x8 in ResHack, it seems to be a bit too much for Metal Gear Solid with CPU+Memory Logic mode.
It's a case of, I would if I could :3. I run on a GTX 260, which may be just about every definition of ancient. It gets me by, quite handily, but I have to keep my expectations in check :3. Until I upgrade, will have to use FRAPs in the mean time =).
So I have an interesting question, is the internal frame rate counter accurate? I ask this because in Blasto, sometimes the frames will drop into the 20's when some heavy action/scenery is going on, but the internal counter will remain showing 59.99/59.98 FPS. Of course I understand why it's 59.99 and not 30, but is it accurate? Does the game just naturally have frame dips during those sections? (I could of course just load the game up on the actual Console, but this would be nice to know in the future for other cases).
@TheDimensioner @Reventon2010 Yeah, I decided to test it out by running really high settings with xBRZ and see if the internal FPS counter adjusted accordingly. xBRZ is a super CPU resource hog, so running 8x12 with 4x xBRZ made the CPU skyrocket to 85% during the first scene in Metal Gear Solid, when you see the submarine in the water.
Obviously my frames dropped from 30 to 15~ and the internal counter went from roughly 59.9 to 30~, so half and half on both of them.
I also tried Blasto at native PSX resolution with no enhancements at all and when the game dropped to 20~ FPS (Via Afterburner), the internal counter stayed at 59.9.... Which tells me the game is just awful and has legitimate drops in frame rate.
Totally off-topic, because xBRZ is such a demanding enhancement (seriously, breaks my poor little CPU (AMD Phenom II X4 3.6 Ghz)) I decided to play around with Hi-Res Textures, specifically 2xSal.
Uhm, totally surprised, because it actually looks good and I can still run ridiculous resolutions in ResHack like 8x12 or 6x6 (Depending on game and how anal I am for wanting perfect frames). It's not as good as xBRZ obviously, but it has this neat effect where it'll leave certain text untouched, meaning it's still super pixel-y sharp whilst 3D elements and things will get filtered with it. For now, it's a pretty good alternative to xBRZ, I'll use it since it's much easier to play games in really high resolution. At 8x12,8x8 and 6x6, the 3D elements in Xenogears (The Gears mostly) are super sharp and have next to no aliasing at all, which is... well freaking awesome, looks sharper and better than xBRZ because I can turn the resolution up so high.
So yeah, that was fun and thanks for the explanation Dimensioner, I knew it wasn't really an FPS counter, instead it had some odd exception, but it is tied to the actual framerate of the game.
Obviously my frames dropped from 30 to 15~ and the internal counter went from roughly 59.9 to 30~, so half and half on both of them.
I also tried Blasto at native PSX resolution with no enhancements at all and when the game dropped to 20~ FPS (Via Afterburner), the internal counter stayed at 59.9.... Which tells me the game is just awful and has legitimate drops in frame rate.
Totally off-topic, because xBRZ is such a demanding enhancement (seriously, breaks my poor little CPU (AMD Phenom II X4 3.6 Ghz)) I decided to play around with Hi-Res Textures, specifically 2xSal.
Uhm, totally surprised, because it actually looks good and I can still run ridiculous resolutions in ResHack like 8x12 or 6x6 (Depending on game and how anal I am for wanting perfect frames). It's not as good as xBRZ obviously, but it has this neat effect where it'll leave certain text untouched, meaning it's still super pixel-y sharp whilst 3D elements and things will get filtered with it. For now, it's a pretty good alternative to xBRZ, I'll use it since it's much easier to play games in really high resolution. At 8x12,8x8 and 6x6, the 3D elements in Xenogears (The Gears mostly) are super sharp and have next to no aliasing at all, which is... well freaking awesome, looks sharper and better than xBRZ because I can turn the resolution up so high.
So yeah, that was fun and thanks for the explanation Dimensioner, I knew it wasn't really an FPS counter, instead it had some odd exception, but it is tied to the actual framerate of the game.
@Reventon2010 Ahahaha, we just had this conversation over at the tweaks section of the forum, it's a bit more complicated then that.
However, for your question as to why you don't see benefits, typically (more true with emulators than with DSR or legitimate downscaling) once you reach the render resolution of your monitor, the "downscaling" that should occur becomes really really hard to notice. Emulators, as far as I'm concerned, don't incorporate high quality algorithms like Lanczos or anything for downsampling, it's typically down in Bicubic or Bilinear I believe, which isn't the best to notice the downsampling effects.
Link to where the ResHack discussion "kind-of" beings - http://ngemu.com/threads/peteopengl...gl2-plugin-w-gte-accuracy-hack.160319/page-72
For PSxFX (I use this as well), I found that for FXAA to work better, make sure you enable full screen smoothing. On another note, Differences between 1-4 on FXAA aren't noticeable either.
For the last question, no idea.
However, for your question as to why you don't see benefits, typically (more true with emulators than with DSR or legitimate downscaling) once you reach the render resolution of your monitor, the "downscaling" that should occur becomes really really hard to notice. Emulators, as far as I'm concerned, don't incorporate high quality algorithms like Lanczos or anything for downsampling, it's typically down in Bicubic or Bilinear I believe, which isn't the best to notice the downsampling effects.
Link to where the ResHack discussion "kind-of" beings - http://ngemu.com/threads/peteopengl...gl2-plugin-w-gte-accuracy-hack.160319/page-72
For PSxFX (I use this as well), I found that for FXAA to work better, make sure you enable full screen smoothing. On another note, Differences between 1-4 on FXAA aren't noticeable either.
For the last question, no idea.
Just wanted to show a picture of Crash 2, wanted to take an image in the same area you did, but... well I didn't want to play until I got there :3.
Anyhow, only difference is I do have 2xSAL high-res textures enabled alongside 8x8 ResHack and FXAA from Asmodeans PSxFX shaders.
![]()
Anyhow, only difference is I do have 2xSAL high-res textures enabled alongside 8x8 ResHack and FXAA from Asmodeans PSxFX shaders.
@Reventon2010 Here is a picture of my settings, the only thing that matters will be the boxed in location.
http://i.picpar.com/svac.png
![]()
Here is what my PSxFX settings look like, you can copy them 1-1 if you'd like, I also have Phong Shading and Color Correction on, they have benign effects on performance, but the slight difference that they give the image can be noticed.
http://pastebin.com/9s8tEdHj
It should work like that. FXAA isn't perfect and you will notice aliasing in the image with certain angles, but most of the time, you should have really nice lines from anything 3D based.
Here's what the game looks like with no filtering, no high-res textures and xBRZ disabled.
http://i.picpar.com/0vac.jpg
![]()
http://i.picpar.com/svac.png
Here is what my PSxFX settings look like, you can copy them 1-1 if you'd like, I also have Phong Shading and Color Correction on, they have benign effects on performance, but the slight difference that they give the image can be noticed.
http://pastebin.com/9s8tEdHj
It should work like that. FXAA isn't perfect and you will notice aliasing in the image with certain angles, but most of the time, you should have really nice lines from anything 3D based.
Here's what the game looks like with no filtering, no high-res textures and xBRZ disabled.
http://i.picpar.com/0vac.jpg
I run a GTX 260, and I don't believe the entire shader would stop working just because your missing that feature.
If you still can't get FXAA working with the settings above, set that line to disable and see if anything happens/changes. (Unless it's already disabled automagically)
Note: I decided to check up on that feature, here is a page I found that suggests your GPU enables that specific OpenGL function, so there shouldn't be any issues on your end. http://feedback.wildfiregames.com/report/opengl/feature/GL_ARB_shading_language_420pack
http://feedback.wildfiregames.com/report/opengl/device/GeForce 9500 GT - For your GPU in specific
Hope this helps you!
If you still can't get FXAA working with the settings above, set that line to disable and see if anything happens/changes. (Unless it's already disabled automagically)
Note: I decided to check up on that feature, here is a page I found that suggests your GPU enables that specific OpenGL function, so there shouldn't be any issues on your end. http://feedback.wildfiregames.com/report/opengl/feature/GL_ARB_shading_language_420pack
http://feedback.wildfiregames.com/report/opengl/device/GeForce 9500 GT - For your GPU in specific
Hope this helps you!
That was quite enjoyable to watch and read @TheDimensioner , I too would love to see some updates for PGXP, keep checking in daily :3.
It's very possible that PCSX-r and PGXP or epsxe and it's GTE stuff is 'hammering' your computer in a way that those benchmarks and stress tests don't, which results in a crash for your computer.
I'd recommend trying to take things back down to stock for a simple test; ascertain if the crashes stop in such an environment.
If it still crashes.... then.... no idea; it's a little strange to see both programs crash with these enhancements turned on.
@CloudXIV Yeah, I've had some software (games) that simply didn't like my computer being overclocked how it was, so returning it to stock or taking it down a few notches solved that issue, I would honestly recommend just trying it. RAM and CPU overclocks back down to stock and see how it goes.
TheDimensioner also brings up some good tips to try as well.
TheDimensioner also brings up some good tips to try as well.
Hellow everyone! I bring with me two videos to showcase PGXP!
V-Rally 2 (NFS V Rally 2 or V Rally 2: Championship Edition for EU folks) and Bloody Roar 2.
It took a while to get to the stage I wanted to show in V-Rally 2, the game is legitimately difficult in Expert mode giving you only 3 retries for the entire arcade race of 6 tracks and there are checkpoints as well. The driving in this game feels exhilirating and really fantastic, I'm surprised such a game exists on the PlayStation and aside from the terrible LOD of the game, it looks really pretty. I'm not sure if the light behaving the way it did in the video is how it looks like on a real PlayStation, I should try software mode or my PS1 and see how it looks like there, but it's pretty neat anyhow.
Bloody Roar 2 is.... amazing, I loved this fighter as a kid and it holds up so well, even without the enhancements of todays technology.
Details on my settings (or some of them) are in the description of each video.
I'm also working on another video to record and upload should the results turn out well. Oh and the 20 minute Bloody Roar 2 video? Yeah, that thing was 30GB's raw, took it down to 1.6 GB's at probably very placebo encoding settings, but we all know the YouTube loves to compress things to hell and back.
Resolution on the Bloody Roar video is at 960x720, couldn't run it at my normal resolution of 1360x1020, I was running at 40~ FPS instead of the 60 required.
Anyhow, ignore my terrible fighting and driving, enjoy the PGXP awesome-ness instead.
- V Rally 2
- Bloody 2
V-Rally 2 (NFS V Rally 2 or V Rally 2: Championship Edition for EU folks) and Bloody Roar 2.
It took a while to get to the stage I wanted to show in V-Rally 2, the game is legitimately difficult in Expert mode giving you only 3 retries for the entire arcade race of 6 tracks and there are checkpoints as well. The driving in this game feels exhilirating and really fantastic, I'm surprised such a game exists on the PlayStation and aside from the terrible LOD of the game, it looks really pretty. I'm not sure if the light behaving the way it did in the video is how it looks like on a real PlayStation, I should try software mode or my PS1 and see how it looks like there, but it's pretty neat anyhow.
Bloody Roar 2 is.... amazing, I loved this fighter as a kid and it holds up so well, even without the enhancements of todays technology.
Details on my settings (or some of them) are in the description of each video.
I'm also working on another video to record and upload should the results turn out well. Oh and the 20 minute Bloody Roar 2 video? Yeah, that thing was 30GB's raw, took it down to 1.6 GB's at probably very placebo encoding settings, but we all know the YouTube loves to compress things to hell and back.
Resolution on the Bloody Roar video is at 960x720, couldn't run it at my normal resolution of 1360x1020, I was running at 40~ FPS instead of the 60 required.
Anyhow, ignore my terrible fighting and driving, enjoy the PGXP awesome-ness instead.
Both games have their textures being filtered, I thought V-Rally 2 looked quite good.
I'll record another video of a different track, but I'll upload it to a cloud storage site so you can download it and watch it without any compression or anything.
EDIT: Here's a zip of two files, one without and one with enhancements. The file without enhancements has everything turned off, no xBRZ, no Full-screen smoother, no FXAA from Asmodeans Shader and texture filtering is set to 0.
The beauty of emulation is you can pick and choose how you want the game to look.
The enhancements has texture filtering set to 6 (Extended with Sprites), FXAA, Full-screen Smoother, but, it does not have xBRZ. I turned off the text in the upper right hand corner (that's how I'd typically play anyhow).
You're free to like what you like, but I think this game, and many others, look better with filtering and sometimes with xBRZ, which is quite subjective.
Link - https://drive.google.com/open?id=0By9ra_2h_FGFTmx6cTRpR2Zfa1U
@TheDimensioner I use handbrake to encode my videos, h.264 at slow, CRF of 14, film tuned (because..... I think that's okay in this circumstance). I don't change or play with resolution and in PCSX-R, I use the keep aspect ratio, stretch on small heights setting, but I keep my resolution 4:3 at all times. Since these games are 4:3 originally, I don't use the widescreen hack or anything like that. For some reason, YouTube didn't want to keep my V-Rally 2 video's in 1080p, of which they're much closer to resolution wise, than 720p, so that sucked.
@cloudbill That's really interesting, I'll have to see how that works in combination with FXAA in some games I play.
@cloudbill That's really interesting, I'll have to see how that works in combination with FXAA in some games I play.
1 - 20 of 22 Posts
